Example files for blog post: Getting started with MAX (modular#1834)

Author: shashankprasanna

examples/blogs-videos/max-blogpost-demos/build_run_docker.sh

@@ -0,0 +1,23 @@
+# Create the model repository and download ResNet-50 model
+MODEL_REPOSITORY=model-repository
+MODEL_NAME=resnet50
+SAVED_MODEL_DIR=resnet50_saved_model
+
+mkdir -p $MODEL_REPOSITORY/$MODEL_NAME/1
+cp -a $SAVED_MODEL_DIR $MODEL_REPOSITORY/$MODEL_NAME/1/
+
+# Create Triton config
+cat >$MODEL_REPOSITORY/$MODEL_NAME/config.pbtxt <<EOL
+instance_group {
+ kind: KIND_CPU
+}
+default_model_filename: "$SAVED_MODEL_DIR"
+backend: "max"
+EOL
+
+# run the recently built max_serving_local container
+docker run -it --rm --network=host \
+ -v $PWD/$MODEL_REPOSITORY/:/models \
+ public.ecr.aws/modular/max-serving-de \
+ tritonserver --model-repository=/models --model-control-mode=explicit \
+ --load-model=$MODEL_NAME

examples/blogs-videos/max-blogpost-demos/infer-max-serving.py

@@ -0,0 +1,43 @@
+
+import numpy as np
+import tritonclient.http as httpclient
+from PIL import Image
+import tensorflow as tf
+import numpy as np
+import tensorflow as tf
+
+### Triton client ###
+client = httpclient.InferenceServerClient(url="localhost:8000")
+
+### Image pre-processing ###
+def image_preprocess(img):
+   img = np.asarray(img.resize((224, 224)))
+   img = np.stack([img])
+   img = tf.keras.applications.resnet50.preprocess_input(img)
+   return img
+
+### Image to classify ###
+img= Image.open('max/examples/inference/resnet50-python-tensorflow/input/leatherback_turtle.jpg')
+img = image_preprocess(img)
+
+### Inference request format ###
+inputs = httpclient.InferInput("input_1",
+                              img.shape,
+                              datatype="FP32")
+inputs.set_data_from_numpy(img, binary_data=True)
+
+outputs = httpclient.InferRequestedOutput("predictions",                                         binary_data=True,                                         class_count=1000)
+
+### Submit inference request ###
+results = client.infer(model_name="resnet50",
+                      inputs=[inputs],
+                      outputs=[outputs])
+inference_output = results.as_numpy('predictions')
+
+### Process request ###
+idx = [int(out.decode().split(':')[1]) for out in inference_output]
+probs = [float(out.decode().split(':')[0]) for out in inference_output]
+
+### Decoding predictions ###
+probs = np.array(probs)[np.argsort(idx)]
+print(tf.keras.applications.resnet.decode_predictions(np.expand_dims(probs, axis=0), top=5))

examples/blogs-videos/max-blogpost-demos/max-engine-resnet50.py

@@ -0,0 +1,37 @@
+
+import shutil
+import tensorflow as tf
+import numpy as np
+from tensorflow.keras.applications.resnet50 import ResNet50
+from PIL import Image
+
+def load_save_resnet50_model(saved_model_dir = 'resnet50_saved_model'):
+   model = ResNet50(weights='imagenet')
+   shutil.rmtree(saved_model_dir, ignore_errors=True)
+   model.save(saved_model_dir, include_optimizer=False, save_format='tf')
+saved_model_dir = 'resnet50_saved_model'
+load_save_resnet50_model(saved_model_dir)
+
+#============================================#
+### MAX Engine Python API ###
+from max import engine
+sess = engine.InferenceSession()
+model = sess.load('resnet50_saved_model')
+#============================================#
+
+def image_preprocess(img, reps=1):
+   img = np.asarray(img.resize((224, 224)))
+   img = np.stack([img]*reps)
+   img = tf.keras.applications.resnet50.preprocess_input(img)
+   return img
+
+img= Image.open('max/examples/inference/resnet50-python-tensorflow/input/leatherback_turtle.jpg')
+img = image_preprocess(img)
+
+### MAX Engine Python API ###
+#============================================#
+outputs = model.execute(input_1=img)
+#============================================#
+
+probs = np.array(outputs['predictions'][0])
+print(tf.keras.applications.resnet.decode_predictions(np.expand_dims(probs, axis=0), top=5))